|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Randomized Trial of Adjuvant Chemotherapy With Mitomycin, Fluorouracil, and Cytosine Arabinoside Followed by Oral Fluorouracil in Serosa-Negative Gastric Cancer: Japan Clinical Oncology Group 9206–1
From the Department of Surgery, Niigata Cancer Center Hospital, Niigata; Department of Surgery, Cancer Institute Hospital; Department of Surgery, Tokyo Komagome Metropolitan Hospital; Department of Surgery, National Cancer Center Hospital, Central; Department of Surgery, National Cancer Center, Research Institute, Tokyo; Department of Surgery, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka; Department of Surgery, National Cancer Center Hospital, East, Tokyo; Department of Surgery, Aichi Cancer Center, Nagoya; Department of Surgery, National Sendai Hospital, Sendai; and Department of Surgery, Kanagawa Cancer Center, Yokohama, Japan. Address reprint requests to A. Nashimoto, MD, Department of Surgery, Niigata Cancer Center Hospital, 2-15-3 Kawagishicho, Niigata 951-8566, Japan; email: nasimoto{at}niigata-cc.niigata.niigata.jp.
Purpose: To evaluate the survival benefit of adjuvant chemotherapy after curative resection in serosa-negative gastric cancer patients (excluding patients who were T1N0), we conducted a multicenter phase III clinical trial in which 13 cancer centers in Japan participated. Patients and Methods: From January 1993 to December 1994, 252 patients were enrolled into the study and allocated randomly to adjuvant chemotherapy or surgery alone. The chemotherapy comprised intravenous mitomycin 1.33 mg/m2, fluorouracil (FU) 166.7 mg/m2, and cytarabine 13.3 mg/m2 twice weekly for the first 3 weeks after surgery, and oral FU 134 mg/m2 daily for the next 18 months for a total dose of 67 g/m2. The primary end point was relapse-free survival. Overall survival and the site of recurrence were secondary end points. Results: Ninety-eight percent of patients underwent gastrectomy with D2 or greater lymph node dissection. There were no treatment-related deaths and few serious adverse events. There was no significant difference in relapse-free and overall survival between the arms (5-year relapse-free survival 88.8% chemotherapy v 83.7% surgery alone; P = .14 and 5-year survival 91.2% chemotherapy v 86.1% surgery alone; P = .13, respectively). Nine patients (7.1%) in the chemotherapy arm and 17 patients (13.8%) in the surgery-alone arm had cancer recurrence. Conclusion: There was no statistically significant relapse-free or overall survival benefit with this adjuvant chemotherapy for patients with macroscopically serosa-negative gastric cancer after curative resection, and there was no statistical difference between the two arms relating to the types of cancer recurrence. We do not recommend adjuvant chemotherapy with this regimen for this population in clinical practice.
THE MAIN aim of the adjuvant chemotherapy for curatively resected gastric cancer is to prevent distant recurrence and increase the cure rate of patients. Although gastric cancer with no serosal invasion (S0–1 cancer) has a good prognosis in Japan, about 20% of patients develop recurrence after potentially curative surgery.1 The most frequent causes for distant recurrence of serosa-negative gastric cancer are hematogenous and lymphatic metastases. Liver recurrence is the most frequent form of failure for serosa-negative gastric cancer.2 Recently, the results of a phase III clinical randomized trial of mitomycin (MMC) + fluorouracil (FU) + oral tegafur + uracil (UFT) after curative gastrectomy for macroscopically serosa-negative gastric cancer (Japan Clinical Oncology Group [JCOG] 8801) was reported,3 and it was concluded that patients with T1 cancer could be excluded from future trials because curative surgery alone yielded a high survival proportion, and there seemed to be no need for adjuvant therapy. No definitive conclusions about the efficacy of adjuvant chemotherapy for gastric cancer have been reached. Encouraged by the favorable results with a regimen of intravenous MMC, FU, and cytarabine (Ara-C) followed by oral FU for stage I and II disease,4 the Gastric Cancer Surgical Study Group (GCSSG), a subgroup of JCOG, conducted a prospective, randomized, controlled study of adjuvant chemotherapy with MMC, FU, and Ara-C followed by oral FU for serosa-negative gastric cancer after curative gastrectomy. In this study, 13 cancer centers participated. We report the results at median follow-up of 69 months.
Patients From January 1993 to December 1994, 252 patients were enrolled in this phase III study. Patients had to fulfill the following eligibility criteria: histologically confirmed adenocarcinoma of the stomach; N2 or less lymph node metastasis; macroscopically serosa-negative (S0, no serosal invasion; S1, suspected serosal invasion) cancer resected without residual disease, excluding T1 (cancer invasion to mucosa or submucosa) N0 (no lymph node metastasis); age 75 years or younger; an adequate bone marrow function (leukocyte at least 4,000/mm3 and platelets at least 100,000/mm3); adequate liver function (ALT, AST, and total bilirubin no higher than 1.25 times the upper limit of normal range); adequate renal function (blood urea nitrogen and creatinine no higher than 1.25 times the upper limit of normal range); no serious complications equivalent to grade 2 or higher toxicity by JCOG toxicity criteria;5 no concurrent active malignancy; no history of other malignancy; and provision of written informed consent. Patients were allocated randomly to adjuvant chemotherapy or no further treatment after curative resection. Pathologic specimens were classified as differentiated and undifferentiated carcinoma.
Treatment Assignment and Evaluations The chemotherapy comprised intravenous MMC 1.33 mg/m2, FU 166.7 mg/m2, and Ara-C 13.3 mg/m2 twice weekly for the first 3 weeks after surgery, and oral FU 134 mg/m2/d for the following 18 months (total maximum dose 67 g/m2). Usual blood analyses were carried out before each cycle of intravenous treatment. A full blood count was performed every week to assess hematologic toxicity. During the oral administration of FU, each patient was asked to visit the hospital every 2 weeks, and received a physical examination and laboratory check regularly. Patients underwent upper gastrointestinal series, ultrasonography, computed tomography, or other investigations either as required or every 6 months to confirm the evidence of recurrence. Adverse events were recorded according to the JCOG toxicity criteria.5 Some of the drug adverse effects, such as slight liver damage, may occur even without chemotherapy after surgery; all categories regarded as drug adverse effect were checked also for those patients in the surgery-alone arm for comparison. For the surgery-alone arm, data for these adverse events (except postoperative morbidity and mortality) were collected at the final analysis. The surgery-alone arm received no additional therapy after surgery unless the patient developed a recurrence. The main prognostic factors including age, sex, the extent of serosal and nodal spread, the method and extent of surgery, and histopathologic findings were described according to the general rules issued by the Japanese Research Society for Gastric Cancer Study.6
Study Design and Statistical Analyses Relapse-free survival was measured from the date of random assignment of treatment to the date of the first observation of relapse or the date of death from any cause. If no progression was reported and if the patient had not died, data on relapse-free survival were censored as of the date that the absence of relapse was confirmed. Overall survival was measured from the date of random assignment of treatment to the date of death or the date of the last follow-up. Relapse-free survival and overall survival curves were calculated by the Kaplan-Meier method and compared by the stratified log-rank test with the combination of the macroscopic assessment of tumor extent and lymph node metastasis as strata for all the eligible patients on intention-to-treat basis. The analysis for the toxicity was conducted for all of the randomly assigned patients to evaluate all the toxicity observed in the subjects who received treatments. All the analyses were conducted by SAS software (Version 6.12, SAS Institute, Cary, NC).
Of 252 patients enrolled, one patient in each arm was ineligible. One patient was judged to be ineligible because he was mistakenly enrolled after closure of accrual, and another was ineligible because of age (random assignment of treatment was performed just a few days after the patient’s 76th birthday). Therefore, 127 of 128 patients in the chemotherapy arm and 123 of 124 patients in the surgery-alone arm were eligible. All patients in the surgery-alone arm and 80 patients in the chemotherapy arm completed the prescribed treatment. Forty-seven patients in the chemotherapy arm had incomplete chemotherapy because of disease progression in three patients, toxicity in 21 patients, refusal during the treatment in seven patients, protocol violation in four patients, intercurrent death in one patient, and other reasons in 11 patients.
Distribution of the main prognostic factors across the two groups was well balanced (Table 1
Adverse events were generally mild; the frequency of toxic effects of JCOG grade 3 or higher is listed in Table 2  . There were three deaths within 30 days after treatment ended in both arms: one patient died of methicillin-resistant Staphylococcus aureus colitis 8 days after random assignment of treatment (surgery-alone arm); one patient died of hepatic metastasis 29 days after chemotherapy ended and 193 days after random assignment of treatment (chemotherapy arm); and one patient died of rupture of cerebral aneurysm 10 days after chemotherapy ended and 302 days after random assignment of treatment (chemotherapy arm).
The frequency of postoperative morbidity and mortality was low and is shown in Table 3 . There were no significant differences between the groups. Twenty-seven patients (22.0%) in the surgery-alone arm and 32 patients (25.2%) in the chemotherapy arm had at least one postoperative complication (leakage, pancreatic fistula, peritoneal abscess, pneumonia, other infections, stomal stenosis, ileus, second surgery, and hospital death).
At median follow-up of 69 months, 21 patients in the surgery-alone arm and 13 patients in the chemotherapy arm had died. There was no significant difference in relapse-free (Fig 1 ) and overall survival (Fig 2) between the arms (5-year survival 88.8% [95% confidence interval (CI), 83.2% to 94.3%] in chemotherapy v 83.7% [95% CI, 77.1% to 90.2%] in surgery alone; P = .14, and 91.2% [95% CI, 86.2% to 96.2%] chemotherapy v 86.1% [95% CI, 79.9% to 92.2%] surgery alone; P = .13, respectively). The results for relapse-free survival were not substantially changed after excluding macroscopic type 4 patients or after adjustment for sex by Cox proportional hazards regression.
Subgroup analyses of the main prognostic factors, such as the extent of serosal and nodal spread, the macroscopic type, pathologic stage, and histologic type in the relapse-free survival and overall survival, also revealed no significant differences between the two arms, but the undifferentiated pathologic type showed better overall survival in the chemotherapy arm than in the surgery-alone arm (5-year survival 95.9% chemotherapy v 85.8% surgery alone; P = .04; Fig 3 ).
During follow-up, 17 (13.8%) patients in the surgery-alone arm and nine patients (7.1%) in the chemotherapy arm had a cancer recurrence (Table 4 ). All of these patients underwent chemotherapy after recurrence. There were no differences between the two arms in the types of recurrence, but hematogenous metastasis (six patients, 4.7%) was the most common type in the chemotherapy arm, whereas peritoneal dissemination (seven patients, 5.7%) was the most common in the surgery-alone arm.
No definitive conclusion has yet been drawn from randomized clinical trials of adjuvant chemotherapy for gastric cancer because few studies have shown a positive effect on survival as compared with surgery alone. Meta-analysis is a way of providing the cumulative evidence from several clinical trials. In 1993, Hermans et al7 published the results of meta-analysis of 11 randomized trials in which postoperative adjuvant chemotherapy for gastric cancer was compared with surgery alone. As a result, they found no definitive improvement in the survival (odds ratio, 0.88; 95% CI, 0.78 to 1.08). A recently published update of their analysis indicates a significant survival benefit (odds ratio, 0.82; 95% CI, 0.68 to 0.97).8 Nakajima et al9 indicated a significant survival benefit for MMC-based adjuvant chemotherapy after curative resection compared with surgery alone as a result of meta-analysis of six randomized trials (odds ratio, 0.63; 95% CI, 0.51 to 0.79). Recently, Earle10 evaluated the effect of adjuvant chemotherapy in gastric cancer according to 13 randomized controlled trials in non-Asian countries. They found a significant survival benefit for patients with adjuvant chemotherapy compared with patients with surgery alone (odds ratio, 0.80; 95% CI, 0.66 to 0.97). Subgroup analyses indicated a trend toward a larger effect when analysis was restricted to trials in which at least two thirds of patients had node-positive disease. Mari et al11 presented results of meta-analysis of 20 randomized trials with a surgery-alone arm. They also demonstrated the survival impact of adjuvant chemotherapy on curative surgery for gastric cancer (hazard ratio, 0.82; 95% CI, 0.75 to 0.89). These meta-analyses indicate that a small but definite survival advantage of adjuvant chemotherapy after curative surgery for gastric cancer, or an even larger advantage for some subgroups of patients, can exist when effective chemotherapeutic regimens with sufficient dose-intensity are used. The combination therapy of MMC, FU, and Ara-C (MFC)12,13 produced beneficial effects and favorable results.14 MFC was reportedly effective for advanced gastric cancer, and it a synergistic effect of the three drugs was observed.12 Adjuvant MFC combination therapy followed by oral FU therapy had a trend to a better survival in the subgroup of stage I to III patients (17% difference in 5-year survival; P = .09)4 in our former trial, which compared three chemotherapeutic regimens (MFC+FU, MFC+UFT, and MF+UFT). The 5-year survival proportion of each regimen was 70.8%, 62.5%, and 66.7%, respectively. A prospective randomized controlled trial with a surgery-alone arm was therefore planned to prove the efficacy of MFC+FU. In designing this study, we set a 15% difference in 5-year survival between the arms as clinically significant, which could be modestly and reasonably expected on the basis of the results from previous studies.4,14 To prove significance in the 15% difference in 5-year survival (85% for the chemotherapy and 70% for the surgery-alone arm) with two-sided type I error of 0.05 and type II error of 0.2, a 2-year accrual time, and 5-year follow-up, a sample size of 110 or more patients per arm was necessary. The number of patients actually enrolled (252 patients) was sufficient for us to detect the planned difference. However, the observed survival difference was smaller than that planned, partly because of much better than expected prognosis of the surgery-alone arm; 5-year relapse-free survival was 88.8% in the chemotherapy arm and 83.7% in the surgery-alone arm, and overall survival was 91.2% in the chemotherapy arm and 86.1% in the surgery-alone arm. With the overall survival of 86.1% in the surgery-alone arm, a 15% improvement in survival because of chemotherapy is clearly impossible. Our study may not have sufficient statistical power for detecting smaller, yet still clinically significant survival benefits from this chemotherapy. Considering its low toxicity, this regimen could be a candidate for future trials to detect smaller clinically meaningful differences for these patients. However, it is also important to find subgroups for which the efficacy of the chemotherapy is expected. In our study, patients with S0 cancer had a good prognosis irrespective of their nodal state or chemotherapy. This finding strongly indicates that most of them had no residual tumor after surgery, and that they were likely to have been cured by surgery alone. These patients accounted for 74.8% of the total, and inclusion of their data may have reduced the overall survival difference between the chemotherapy and surgery-alone arms. This finding indicates retrospectively that our entry criteria were not appropriate in this regard. Subgroup analysis of past trials indicated that adjuvant chemotherapy might show a survival advantage for moderately advanced gastric cancer, such as stages II or III. Chemotherapy and concurrent radiation therapy (FU + leucovorin + irradiation) suppressed locoregional relapse and showed a significant prolongation of the median survival.15 This combined-modality therapy was well tolerated and the relapse-free survival was significantly improved. In this trial, however, lymph node dissection was inadequate for enrolled patients with less than D0 for 54% of the patients, D1 for 36% of the patients, and D2 for only 10% of the patients. Considering the high frequency of lymph node metastasis to the first and second nodal stations, contribution of radiotherapy to local control should be enormous. The frequency of D2 or more dissection in our study was 98% and the morbidity and mortality rate was low. For the patients who underwent a curative surgery with D2 or more extended dissection, the necessity of radiation therapy to improve local control is doubtful. The cost per patient of this adjuvant chemotherapy can be calculated theoretically. The full course chemotherapy costs 604,497 yen ($5,600). The intravenous infusion of MFC costs only 9,966 yen (1.6%) and oral FU for the following 18 months costs 594,531 yen (98.4%). A clinically significant survival benefit has not been shown in our study, but the results of the study have shown a possible 5% improvement in 5-year survival by the chemotherapy, with the cost per patient of approximately 600,000 yen ($5,600) for the chemotherapy. A low incidence of adverse events and relatively high compliance were observed. In conclusion, there was no statistically significant relapse-free or overall survival benefit with this adjuvant therapy regimen for patients with macroscopically serosa-negative gastric cancer after curative gastrectomy with D2 or more extended lymph-adenectomy, and there were no major differences between the two arms regarding the types of cancer recurrence. A more than 10% improvement in 5-year survival for this population is unrealistic because of the good prognosis after surgery alone. Studies of adjuvant chemotherapy expecting more than a 10% improvement of 5-year survival for gastric cancer after curative resection should be focused on T3 or a more advanced stage of disease. For serosa-negative patients, large clinical trials to detect smaller but clinically meaningful improvement of survivals with low-toxicity regimens such as MFC+FU are needed.
All investigators contributed to overall study design and coordination. S. Yamamoto performed the statistical analyses. All investigators contributed to the text of the first draft of the paper. A. Nashimoto and S. Yamamoto wrote the final version of the paper. Other members of Gastric Cancer Surgical Study Group include: Kuniyoshi Arai, Department of Surgery, Tokyo Komagome Metropolitan Hospital; Keiichiro Ohta, Department of Surgery, Cancer Institute Hospital; Takeshi Sano, Department of Surgery, National Cancer Center Hospital, Central, Tokyo; Massahiro Hiratsuka, Department of Surgery, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka; Juei Sasaki, Department of Surgery, Niigata Cancer Center Hospital, Niigata; Akira Kurita, Department of Surgery, Shikoku Cancer Center National Hospital, Matsuyama; Tsuneaki Fujitani, Department of Surgery, Miyagi Cancer Center, Sendai; Kenji Uchida, Department of Surgery, Saitama Cancer Center, Saitama; and Ken Kondo, Department of Surgery, National Nagoya Hospital, Nagoya, Japan.
We thank Harumi Kaba for data management and Naoki Ishizuka, MD, for the assistance of statistical analysis; Haruhiko Fukuda, MD, Director of the JCOG data center, for his helpful comments to the manuscript; B.G. Mann, MD, for his English corrections and suggestions; and all of the colleagues taking part in this study. Without their support this trial could not have been performed.
Supported by Grants-in-Aid for Cancer Research (2S-1, 5S-1, 8S-1, 11S-3, and 11S-4) from the Ministry of Health, Labor and Welfare.
1. Nakajima T: Tabular analysis of l0,000 cases of gastric cancer. Jpn J Cancer Chemother 21:1813–1897, 1994 2. Sano T, Kobiri O, Muto T: Recurrence of early gastric cancer. Stomach and Intestine 28:173–176, 1993 3. Nakajima T, Nashimoto A, Kitamura M, et al: Adjuvant mitomycin and fluorouracil followed by oral uracil plus tegafur in serosa-negative gastric cancer: A randomized trial. Lancet 354:273–277, 1999[CrossRef][Medline] 4. Nakajima T, Takahashi T, Takagi K, et al: Comparison of 5-fluorouracil with ftorafur in adjuvant chemotherapies with combined inductive and maintenance therapies for gastric cancer. J Clin Oncol 2:1366–1371, 1984[Abstract]
5. Tobinai K, Kohno A, Shimada Y, et al: Toxicity grading criteria of the Japan Clinical Oncology Group. Jpn J Clin Oncol 23:250–257, 1993 6. Japanese Research Society for Gastric Cancer: The General Rules for the Gastric Cancer Study (ed 11). Tokyo, Japan, Kanehara & Co, 1985 7. Hermans J, Bonenkamp JJ, Boon MC, et al: Adjuvant therapy after curative resection for gastric cancer: Meta-analysis of randomized trials. J Clin Oncol 11:144–147, 1993 8. Hermans J, Bonenkamp JJ: In reply. J Clin Oncol 12:879–880, 1994 (letter) 9. Nakajima T, Ota K, Ishihara S, et al: Meta-analysis of 10 postoperative adjuvant chemotherapies for gastric cancer in CIH. Jpn J Cancer Chemother 21:1800–1805, 1994 10. Earle CC, Maroun JA: Adjuvant chemotherapy after curative resection for gastric cancer in non-Asian patients: Revisiting a meta-analysis of randomized trials. Eur J Cancer 35:1059–1064, 1999[CrossRef][Medline]
11. Mari E, Floriani I, Tinazzi A, et al: Efficacy of adjuvant chemotherapy after curative resection for gastric cancer: A meta-analysis of published randomized trials. Ann Oncol 11:837–843, 2000 12. Ota K, Kurita S, Nishimura M, et a1: Combination therapy with mitomycin C (NSC26980), 5-fluorouracil (NSC19893), and cytosine arabinoside (NSC63878) for advanced cancer in man. Cancer Chemother Rep 56:373–385, 1972[Medline] 13. Cocconi G, Delisi V, Di Blasia B: Randomized comparison of 5-FU alone or combined with mitomycin C and cytarabine (MFC) in the treatment of advanced gastric cancer. Cancer Treat Rep 66:1263–1266, 1982[Medline]
14. Nakajima T, Fukami A, Takagi K, et al: Adjuvant chemotherapy with mitomycin C, and with a multi-drug combination of mitomycin C, 5-fluorouracil and cytosine arabinoside after curative resection of gastric cancer. Jpn J Clin Oncol 10:187–194, 1980 15. Nakajima T, Okabayashi K, Nakazato H, et al: Effect of MFC-based adjuvant chemotherapy in gastric cancer with curative surgery. Jpn J Soc Cancer Ther 29:654–662, 1994
16. Macdonald JS, Smalley SR, Benedetti J, et al: Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med 345:725–730, 2001 Submitted June 18, 2002; accepted January 21, 2003.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
